Expert: Robert Cummings, P.E. - 2/23/2008

Question
Mr. Cummings:
This is a followup to the 48'x24' retaining wall failing post on February 19.
I will try to answer the  the points you raised to clarify the underlying issues from the first post.(I saw no way to add to the original post).
1.The wall does retain backfill. And after further examination the concrete blocks are 12"Wx16"L(the end blocks are 8"W x 16"L)-not 8"x16"L as I said previously.I carefully removed part of one. The backfill is dry sand and does have metal mesh running through.
2.The garage walls rest on the retaining wall and both sides. The top row of block is "L" block and it appears to be filled with concrete.(Paper was jammed on the row of blocks below the top row to prevent concrete filling all blocks-I do not know why.).It appears that there was no expansion joint put in from the slab to the "L" block. When I look  at the top block(behind the vinyl siding), I see the sill plate.Perhaps they shot dowels through the sill plate to the concrete filling the "L" block. It appears that the outside lower edge of the top row of blocks has lifted from the block below(most pressure is on the inside edge of the top block).
2.The slope on side#1 is 0 until the last 7 feet of block, where it is appx 1' per foot.(this side sees most of the water as a hill  is to this side and to the front corner.
3.There are no open springs on the property but twenty feet downhill from this retaining wall there is a 2-3 foot wide 2 inch stream of flowing water(from a spring on other property. The grade from the retaining wall to this stream is 1 foot drop every 2-3 foot towards the stream.
4.Yes. Cold weather of below 32 degrees does occur appx  1x per week during winter(Dec-Feb), sometimes getting down to 15 degrees for a night.
5.There are no drains in the wall, and the backfill soil is dry sand with small pebbles.(at least where I removed that block). I intend to get a 12" long concrete bit today and drill out some holes on lower courses to determine if water there.
6.Remedial design:The piers would be placed every 6' apart across the back of the retaining wall. I have a Stone tamper(gas powered motor) to use to tamp the soil(after the footer hole is dug) and before the concrete is poured.These piers would be on the downhill side butting up against the wall.
And one pier on each end(wrapping 16" to the side for a 32" L shaped pier). #4 rebar would be put in the footer and 16"Lx16"W blocks would be used.The pier would be bolted the the existing concrete block and not to the existing slab(too long of bolt required to get to existing slab)(ignore the original post about tying the pier to the slab because we found the top block to be "L" block). We do not want to remove the blocks as the fill will come out and the slab will lose support and possibly crack(unless we come up with a way).
The slab is not post tensioned, and the pier would not underpin the slab UNLESS we removed a small section of the back wall and put a floor jack from the footer to the slab then built up the blocks to the slab. The bottom of the slab would be drilled for rebar fill the holes then epoxy and then run rebar from slab to newly poured footer(16"W x 8"D) and then use a sonotube to pour a concrete pier. To pour, the 16" sonotube from the top where it could be pushed back under the slab when full.

I want to the minimum necessary if the wall will not collapse AND the slab does not crack.
I will drill drain holes first.
I am just concerned about the 1-1/2" deflection in the top middle of the retaining wall.

I do not know what you mean when you say "If a wall does not support a structure it can tolerate a lot of deflection without failing"... What structure are you referring to? I assume the garage. And by wall you mean retaining wall.

If you would like to see a photo of the wall in question I could send one but would need an address. Again thank you for your help.

Answer
I answered the followup right after you posted it and apparently that answer was lost.  Aorry.  I will try to reconstruct what I said from memory.

First, my statement that walls can deflect a lot without failing is true of most internally reinforced retaining walls, but that deflection is undesirable if a structure bears on the top of the wall.  Since your case involves a structure on top of the wall I guess we can conclude that it is necessary to stop the deflection.

Since you do not describe a lot of stairstep cracking in the side walls, and because you note that the deflection is in the middle of the rear wall rather than all along its length, and because the top row of block at the rear of the garage is detaching to the outside, it sounds like the center of the wall is rotating outward down the slope.  I'm not completely sure why.  Because the backfill is sand, which should be free-draining, and is reinforced with mesh, the backfill should not be exerting high passive active soil pressures on the wall causing it to rotate.  If on the other hand the mesh that you have seen is only at the top, and the remainder of the soil beneath it is not so reinforced and is wet, there may be an active earth pressure issue.

Piers are not real good at resisting lateral pressures; to resist lateral pressures I would suggest using tiebacks across the rear of the wall instead.  You would employ heavy steel channel horizontally about 3/4 of the way up the wall with holes pre-drilled for the anchors and their bearing plates.  The anchors should be #8, hollow, self-drilling cement-grouted epoxy coated threadbar (http://www.williamsform.com/Ground_Anchors/Hollow_Bar_Ground_Anchors/hollow_bar_...).  The length and spacing of the anchors depends on the wall height and the characteristics of the backfill.  You drill those in place, allow the grout to set up for at least 72 hours, mark and cut the holes in the channel, and bolt the channel over the anchors thereby applying tension.  That should arrest the rotational behavior.

However there seems to suspect a contribution of partial bearing failure or slope creep, both of which can masquerade as a rotational active pressure failure.  One way to tell would be to examine the connection between the rear and sides of the garage retaining wall.  A bearing problem should extend all along the rear of the wall, to one extent or another, and would be expressed as settlement at the top as well as separation widening to the top.  Also, there should be stairstep cracking in the side walls where they intercept the rear wall.  Pure rotation stemming from active pressure would not do this.  You can address the bearing pressure issue with piers but they should bear below the frost line and deep enough to be out of the zone of slope creep.

I would be interested in a photo.  Send it to sgsibob@aol.com